| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
This allows for example to complete file names in markdown images.
E.g.:  (hint: variables in lua already handle this).
|
|
This will break all plugins which currently use Win.viewport.
|
|
|
|
Apparently Github changed the permissions on /usr/local in the
default image so now we need to use `sudo` for luarocks to be able
to install packages.
While I'm here: update the checkout dependency and use
macOS-latest image instead of multiple old images.
|
|
|
|
Currently, there's no mechanism for defining user styles without risk
of collision with default lexer and theme settings.
Very few lexers use more than 10,
let alone all 64 of the allowed styles, so
UI_STYLE_LEXER_MAX - (no. of user defined styles)
allows for a reasonably large number of user-defined styles
before collision becomes a problem.
|
|
|
|
This also involves modifying the style definitions so that they
include sub tags as well. For example command.section in latex
should be mapped to COMMAND_SECTION.
|
|
Caching lexers causes lexer tables to be constructed once and
reused during each HIGHLIGHT event. Additionally it allows to
modify the lexer used for syntax highlighting from Lua code.
This is used for example for the syntax aware spellchecking
performed by the vis-spellcheck plugin.
|
|
vis{,-std}.lua:
* replace removed _TOKENSTYLES with _TAGS
* don't add default styles to the lexer. lexers no longer
define their own styles
filetype.lua: update path detection for scintillua 6.2
|
|
Based on the advice from the upstream maintainer of Scintillua
(gh#orbitalquark/scintillua#87).
|
|
Rather than cherry pick patches from after 6.2 we will just grab
everything as is.
|
|
The old style handling had a lot edge cases where one of the
colours or the attribute wouldn't get applied correctly. This
commit adds a new style_set() method to the Ui which should be
called instead of manually touching a cell's style. This also
means that the Cell struct can be made opaque since all the
handling is now done inside the ui-terminal files.
With this it is now viable to combine the light and dark 16 colour
themes into a single base-16 theme. This theme works very well
with the Linux virtual console and will now be the default theme
regardless of if the terminal supports 256 colours or not. This
should address the common complaints about vis not respecting the
users default terminal colours.
fixes #1151: Theming is sometimes partially applied or ignored
see #1103: terminal no longer has transparency/opacity
see #1040: Transparent background and setting options by default
|
|
Silencing compiler's -Wparentheses warning.
|
|
|
|
|
|
|
|
|
|
wayclip has since commit
https://git.sr.ht/~noocsharp/wayclip/commit/594776c24862 support
for the choice between using primary or regular clipboard. Which
means we can put its support on the same level as other clipboard
utilities.
|
|
Internally vis supports unicode just fine. Instead of relying on
external programs utilize vis' own features.
Thanks to Florian Fischer for the correct regex!
|
|
There is the textadapt module for dealing with Modula 2
by the late Nikalus Wirth and its successor Modula 3 on
https://github.com/paaguti/textadept-modula-lexers. If I want to
just use it, I need to have these types recognized.
|
|
Unfortunately, GNU tr(1) is not Unicode-aware,
so we should use awk(1) instead. See
https://www.pixelbeat.org/docs/coreutils_i18n/ for more on the
situation of the support of Unicode in coreutils.
Apparently, awk is for this better than sed, because Unicode is
consistenly provided on all major versions of awk.
Signed-off-by: Matěj Cepl <mcepl@cepl.eu>
Supersedes: https://lists.sr.ht/~martanne/devel/patches/49113
Supersedes: https://lists.sr.ht/~martanne/devel/patches/49114
|
|
|
|
|
|
Apparently `labels:` should be an empty list and not an empty string
if you want no default labels; my bad.
|
|
|
|
|
|
This allows to complete file names in latex code (e.g. `\include{foo}`).
|
|
Just a version bump, no changes required.
|
|
This hasn't worked in almost a year and even if it did it makes no
sense. Based on my testing lexing takes a couple milliseconds at
most. If it took 1 second (the default value for this option) vis
would be completely unusable.
If people want support for this it should be submitted upstream
and vis will act based on the outcome of that.
closes #1122: lexer no longer obeys redrawtime
|
|
This obviously draws on the alt_name parameter used in scintillua
but rather then passing it all the way up to the chain to
lexers.load() we will just handle it in set_filetype().
8a420ec accidently readded detection for the removed git-commit
lexer which somehow went unnoticed until we readded caching for
the new lexers. Instead of just removing it we can alias to the
diff filetype and only set the colorcolumn for commit messages.
This fixes the incorrect behaviour of adding a colorcolumn to diff
and patch files and thus completely reverts 0cc684f.
|
|
This adds `[not]dim` to the set of accepted theme keywords
|
|
As defined in https://peps.python.org/pep-0484/#stub-files.
|
|
|
|
When a new subprocess is created during an EXIT event of
another subprocess new_process_in_pool will update the
process_pool pointer. Since we use a pointer to a pointer
for iterating all processes during vis_process_tick its
value will be different before executing the event and after
creating the new subprocess. This causes the updated pointer
to be erroneously destroyed and leaves the Process of the
reaped child behind which causes consecutive waitpid calls
to fail with ECHILD.
This is fixed by destroying the proper current subprocess
and updating the iteration pointer accordingly.
Fixes: 0b015320382e74fcb385a46a81304f588ed27f77
|
|
There are probably more things to simplify but at least this makes
it easier to see what exactly is different between `<C-x><C-f>` and
`<C-x><C-o>`.
Some differences were removed:
* whitespace in range is treated the same for both actions
* empty range will expand to files in CWD for both actions
closes #1146: Complete file name and file path swapped in doc
|
|
fixes #1152: `:2x/foo/<cr>` in a file with only one line makes vis
get stuck in an infinite loop
|
|
fixes #1151 (part 2): Set foreground color for matching pair
|
|
fixes #1151: Set foreground color for visual selection
|
|
The view_style function is used to apply styles to ranges of text in
a view. It approaches the starting position where the style should be
applied by iterating the columns in the appropriate line using
this while loop:
while (pos < start && col < width)
pos += line->cells[col++].len;
The while loop will stop at the last character before the range where
the style should be applied.
This works fine until we encounter "empty" cells between the last
cell containing an actual character and the first cell to be styled.
This can happen if the last character before the range to style is
'\t' which gets expanded with empty cells by vis according to the
tabwidth option. Those empty cells get erroneously styled as well.
This is fixed by skipping all empty cells encountered before the
range to style.
fixes #1147: `win:style` styles more terminal cells than expected
|
|
according to POSIX wait(3p) the return status needs to be checked
and the macro WEXITSTATUS(stat_val) should be used to get the actual
return value on a normal exit. POSIX doesn't specify the value of
stat_val on abnormal exit and thus vis_pipe() should just return
-1 as it does for other errors
closes #1130: vis:pipe returns wrong exit status (when non-zero)
Thanks @pippi1otta for the report and suggestion.
|
|
Sourcehut recurses into submodules when cloning the repo for building
so unlike github it uses the version of `test` that is checked into
the repo. This is better behaviour but does mean that the submodule
needs to be updated.
|
|
aka:
"check for EOF before unsetting row, col & line cache in view_coord_get"
"fix bug where visual-line selections after view were considered visible"
These commits have created more bugs then they fix. Reverting them
reintroduces #1074: Slave selection strangled by view cliff.
Fixes #1143: Disappearing selection
|
|
this is a valid test however the upstream fix is incorrect and needs
to be reverted.
|
|
same as last commit, `fstab.lua` shouldn't be matched as `fstab`
|
